Analysis of a Flexible Concrete Arch

Arch bridges have been in existence for thousands of years and many of them are testament to the durability of such structures. There are approximately 70,000 masonry arches in the UK, many of them forming an important part of the UK road and rail network, equating to approximately 40% of the total used within the UK. ‘FlexiArch’ is a pre-cast concrete arch system which was developed in Queen’s University, Belfast. Individual concrete voussoirs precast to the correct taper for a given span and rise, are connected by a polymeric membrane, which allows them to form an arch when lifted. Modern durability issues of rebar corrosion are avoided as the arch system requires no internal reinforcement. This paper investigates the construction and testing of three third-scale FlexiArch models that have been built in the laboratory under a number of variables. Geotechnical tests alongside non-linear finite element analyses have also been done to determine the influence of the granular backfill. 134 ARCH’10 – 6th International Conference on Arch Bridges Figure 1: Components of FlexiArch system 2 DESIGN AND CONSTRUCTION OF THE FLEXI-ARCH SYSTEM Third-scale 5m x 2m (span x rise) arches were constructed in the laboratory. The geometry of the voussoir blocks was calculated from the overall span and circular profile of the arch. Eight moulds were then designed and made for these using ‘Lexan’ plastic, with the addition of solid tubes to create the hollowcore design (Fig.2). These tubes were removed when casting solid voussoirs in the same moulds. 23 hollowcore blocks of depth 66mm and length 333mm were used to construct each arch with a design span of 1.67m and rise of 0.67m. A third-scale concrete mix was used with a 28-day strength in excess of 30N/mm. Hooks were made using rebar and cast into 3 of the blocks for each arch made. Figure 2 : Hollowcore Voussoir Moulds (left); Voussoirs after demoulding (right) After curing, the top of the blocks were roughened with an angle grinder in order to improve the bond with the screed. The voussoir blocks were then laid on a flat construction bed, and the polymeric reinforcement was cut and bonded to the top of the blocks. ERS strain gauges were then attached to the polymeric reinforcement using an epoxy glue. The strain gauges were attached at the midspan, lifting hooks and the end voussoir blocks to enable strain measurement to be recorded at critical locations for both the construction phase and under load testing. Adhesive was then applied to the surface of the voussoirs to provide a sound bond between them and the 13mm top screed thus connecting each voussoir. The top screed caps the geotextile which connects each block together. Vibration was not possible for a screed of this depth, so the top surface was floated and after one hour crack inducers were then scored into the screed (above the edges of each block below) to ensure controlled cracking at the joints during lifting into the arch form. After the screed had cured for a minimum of seven days, the arch was lifted three times and the strain measured in the polymeric reinforcement (due to carrying the self-weight of the arch) during lifting and lowering. The maximum strain in the polymer was found to be 1200μ, well below the capacity of the material. 3 TEST SET-UP AND PROCEDURE The arch was then placed on correctly sloped anchor blocks under a hydraulic load actuator. The formwork was designed and placed with steel on one side and a strong clear ‘Lexan’ plastic on the other side so that deformation and hinge development could be observed during testing. John Bourke, SE Taylor, D Robinson, etc. 135 Deflection transducers and vibrating wire strain gauges were then installed under the arch on the intrados at the midspan, third points and near the abutments. The arch was then subjected (without backfill) to a proof load with a guideline deflection of 5mm. The arch was then backfilled with granular material in equal 100mm layers on both sides of the arch ring to avoid eccentricity and each layer was compacted using a vibrating hammer with a flat attachment. Deflections under the intrados of the arch were recorded during backfilling operations, but were found to be minimal. The complete third-scale arch bridge was then tested, with the load applied across a 150mm wide plate. In total, three arch rings were constructed and tested. Each model was built with the same span to rise but different backfill was used in test arches 2 & 3. The first arch was constructed with a 44mm diameter core in the voussoirs (as shown in Fig.3), the second was constructed with solid voussoirs, and the third arch was constructed also with hollowcore design like the first (Fig.3).